JP2007316905A - Computer system and method for monitoring application program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for generating, when a position where monitoring is to be executed and data to be monitored are designated in a business process level understandable by a business manager or operation manager, a program module for executing the designated monitoring without manual operation, and designating a program code level necessary for inserting the program module to an application program. <P>SOLUTION: In the computer system, a monitoring request described on design information with high abstraction is converted to a description of program mounting level by associating a point or data on the design information with high abstraction with a point or data on a program code by use of a correspondence between components of design information in each stage of an application development process. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The technology disclosed in the present specification relates to a method for monitoring behavior during execution of an application program, and more particularly, to a method for executing setting and control of application program monitoring.

  An information system operation manager or a business manager monitors the status of the system and the business using a log output by the business application. The log includes a system log output in each layer of the information system that executes business applications such as hardware, OS, and application server, and an application log output by the business application. Each layer of the information system has a function of outputting a system log and a function of customizing a log to be output during operation of the system in response to a request from an operation manager or a business manager.

  In order to monitor the status of the business being processed by the information system, an application log is required instead of the system log. The application log required for business status monitoring may change after the start of application operation due to environmental changes such as amendments to the laws surrounding companies. For this reason, it is necessary to add or change the log output by the application after the operation is started.

  The method for realizing the function of outputting the application log differs depending on the implementation form of the application program.

  Recently, in the development of business applications, a model-centric development method has been used. According to this development method, a manager or a design consultant models a business process using a model description language such as Business Process Modeling Notation (BPMN) (see Non-Patent Document 1). The created model is converted and refined step by step into an implementation level model, thereby creating an implementation level specification of the program. And an application program is mounted based on the produced specification (refer nonpatent literature 2).

  FIG. 2 is an explanatory diagram of an application development process and a model diagram created in the process.

  In application development, a business process diagram 301 created using business process modeling notation (BPMN) or the like, a UML class diagram 302 created using a model description language such as Unified Modeling Language (UML), and the like A UML sequence diagram 303 or the like is used.

  Hereinafter, model diagrams used in application development, such as business process diagrams, UML class diagrams, and UML sequence diagrams, are collectively referred to as design information.

  In such an application development process, first, an abstract business process diagram 301 is created by analyzing and modeling a business process. The created business process diagram 301 is detailed in stages. The detailed business process diagram is further refined into the model of the UML class diagram 302 or the UML sequence diagram 303. By repeating the refinement, the abstract application design model is converted into a model at a level close to the implementation of the application program. Finally, the application program code 304 is implemented based on the implementation level model.

  In recent years, business applications are generally constructed by combining components, which are a plurality of small software parts, using an object-oriented language. As basic technologies for constructing such an application, there are J2EE (see Non-Patent Document 3) and .NET. Applications implemented using these platform technologies are executed on an application operation platform called a J2EE application server or .NET application server.

  On the other hand, as a different approach, a Service Oriented Architecture (SOA) approach has been proposed in which each function constituting a business process is implemented as a service, and an application is implemented as a loosely coupled service using a messaging system (non-patent literature). 4).

  In the application implemented by the SOA, the status of the business processed by the application can be monitored by capturing messages between services in the messaging system (see Non-Patent Document 5). Since monitoring by the messaging system can be set independently of the service, application level monitoring can be realized without creating an application, and monitoring contents can be customized during operation. Alternatively, the contents to be monitored can be added or changed by recording all messages exchanged between services and then referring to the necessary messages.

  In the case of a general application that does not take the SOA approach, the behavior of the application cannot be monitored from the outside by monitoring the behavior of the application like the SOA application. Therefore, when adding or changing a log output by an application, it is necessary to create a function for outputting the log individually for each log, and thus the application program must be changed. Therefore, there is a problem that it takes time to implement a log output code according to each log to be output, and there is a problem that it is necessary to stop the operation of the application in order to change the application program.

  As a method for solving the problem that a program needs to be changed in order to add or change a log output code to an application, there is a method using a bytecode instrumentation technique (see Patent Document 1). Byte code instrumentation technology dynamically adds and changes functions during program execution by dynamically rewriting the program code of an application program. Using this technology, it is possible to realize dynamic addition and change of the log output function.

  Patent Document 2 discloses a system that selects a program module that outputs a log prepared in advance using a bytecode instrumentation technique and inserts the program module into a running application program. . As a result, arbitrary application monitoring can be added without stopping application operation.

  Japanese Patent Application Laid-Open No. 2004-228561 discloses a system that monitors execution performance by inserting a program module for executing monitoring into a designated portion of an application program during execution of the application program. In this system, the location of the program to be monitored is input at the program code level. Then, a program module for executing monitoring prepared in advance is automatically inserted at the input position, and monitoring is executed.

  In addition, there is a method of automatically generating a log output code in order to solve the problem that it takes time and effort to implement the log output code. Non-Patent Document 6 shows a system that automatically generates a program module for monitoring an application. In this system, a program module for monitoring an application is automatically generated by inputting a portion to be monitored and data to be monitored at the program code level.

  In addition, as another method for adding or changing the log to be output during the operation of the application, in Patent Document 4, the monitoring function that has been created in advance in the application is selected during the operation of the application. A system with a user interface that controls the execution of monitoring is shown. In this method, execution of monitoring can be controlled during operation within the scope of the monitoring function created before application operation. However, selectable monitoring functions cannot be added or changed during application operation.

  Further, when executing application monitoring, it is necessary to consider performance degradation caused by execution of monitoring. The process of monitoring the application and outputting the monitoring data generates a certain load. Therefore, it is necessary to execute monitoring so that the monitoring load does not affect the execution of the application program as much as possible.

  Some monitoring items do not always need to be monitored, but there are also monitoring items that need to be monitored under specific circumstances. When there are a large number of such monitoring items, the monitoring load increases by constantly executing such monitoring, and the load may greatly affect the execution of the application. In order to cope with this problem, there is a method in which conditions for executing monitoring are set, and monitoring execution is controlled as necessary, instead of always executing all monitoring. Patent Document 5 discloses a system that controls whether or not to acquire another monitoring item according to the value of a certain monitoring item. This control makes it possible to perform necessary monitoring according to the situation without applying an unnecessary monitoring load. In this system, a condition for executing monitoring is set by setting a monitoring data threshold and a monitoring item to be executed when the threshold is exceeded.

In Patent Document 6, the time required for log output is measured in the log output device used by the application for log output, and the time required for log output is less than a certain percentage of the time required for execution of the entire application. A system for adjusting the output level of the log is shown.
US Pat. No. 6,260,187 US Patent Application Publication No. 2005/0273667 JP 2005-523518 A JP 2003-233515 A JP 2004-206495 A JP 2001-175508 A By Stephen A. White, translated by JGC Information Software Co., Ltd., “Outline of BPMN”, [online], May 2004, JGC Information Software Co., Ltd. [searched on December 13, 2005] , Internet <URL: http://www.jsys-products.com/iwaken/bpmn/pub/Introduction_to_BPMN.pdf> Bill Shannon, “Java 2 Platform Enterprise Edition Specification, v1.4” [online], November 24, 2003, Sun Microsystems, Inc. ( Sun Microsystems, Inc.) [Search June 23, 2005], Internet <URL: http://java.sun.com/j2ee/j2ee-1_4-fr-spec.pdf> By David S. Frankel, translated by IBM Japan, TEC-J MDA Subcommittee, “MDA Model Driven Architecture”, SIB Access Inc., November 1, 2003 Japan BEA Systems Co., Ltd., “SOA Service Oriented Architecture”, Shosuisha Co., Ltd., March 22, 2005 David Chappell, "Understanding BizTalk Server 2004," [online], October 2003, Microsoft Corporation, [searched June 23, 2005], Internet <URL: http://download.microsoft .com / download / 1/1/1 / 1118730e-ec93-45aa-8c06-97af628db61d / UnderstandingBizTalkServer2004_J.doc> AdventNet, Inc., “Manage Engine JMX Studio 5 Product Documentation” [online], 2004, AdventNet, Inc., 2005 Search February 2], Internet <URL: http://manageengine.adventnet.com/products/jmx_studio/AdventNetManageEngineJMXStudio.pdf>

  The conventional techniques for monitoring these application programs have the following problems.

  First, in order to add or change a monitoring function to an application program, the conventional technique automatically generates a program module for executing monitoring, and applies the generated program module during execution of the application. However, in order to use such a conventional technique, it is necessary to input the location (point) for monitoring and the data to be monitored at the program code level.

  In general, it is often an operation manager or a business manager that needs a log to be added or changed. In general, operation managers and business managers do not understand application program code. Therefore, the operation manager and the business manager cannot input the contents to be monitored at the program code level necessary for using the conventional technology. Therefore, it is necessary for an application designer or programmer who understands the program code to create and input necessary information in response to a request from the operation manager or business manager. There is a problem in that this labor is an obstacle to the addition or change of the rapid monitoring function.

  In order to cope with this problem, it is possible to prepare information to be input in advance, and apply the information selected by the operation manager as needed from the information. However, it is difficult to grasp in advance all the monitoring contents that may be necessary and prepare necessary input information. Therefore, the above method cannot deal with a case where it is necessary to add monitoring contents that were not assumed in advance, and is not an essential solution.

  Second, by using a conventional technique, it is possible to control items to be monitored according to the state of the application. Here, when a condition for controlling execution of monitoring is specified, a monitoring item used for determining the condition is specified. Furthermore, a monitoring item to be controlled according to the state is designated. At this time, if the monitor items that can be specified are named at the program code level, the monitoring control cannot be set appropriately unless the program code is understood. In addition, when there is no function for monitoring data to be used for condition determination, it is necessary to newly add a monitoring function.

  In the prior art, the monitoring execution is controlled by filtering the output of the monitoring data from the application. Therefore, the program code for monitoring is incorporated in the application program regardless of whether or not monitoring data is output, and processing of the monitoring code is executed according to the execution of the application. Therefore, even if the output of monitoring data is controlled, the monitoring load cannot be reduced completely.

  Third, it is generally difficult to estimate the monitoring load without understanding the program implementation. If the first problem is solved, the application monitoring function can be added or changed without being aware of the implementation of the application program. However, in this case, unintentionally adding a monitoring function with a high monitoring load to the application program may cause a serious performance degradation in the execution of the application program.

  In the prior art, since the monitoring load is measured in the log output device outside the application, it is not possible to measure the load of monitoring data acquisition, generation and output request processing to be processed inside the application program. . In the prior art, the load of the entire monitoring function of the application is measured, and the execution of monitoring is controlled based on the result. For this reason, the load of the individual monitoring function is measured, and the execution of monitoring cannot be controlled based on the result.

  From the above points, the first problem to be solved by the present invention is that when a location to be monitored and data to be monitored are designated at a business process level that can be understood by a business manager or an operations manager, It is an object of the present invention to provide a method for generating a program module for executing specified monitoring without manual intervention and specifying a program code level necessary for inserting the program module into an application program.

  The second problem to be solved by the present invention is to control application monitoring based on conditions, thereby enabling appropriate monitoring according to the situation without requiring knowledge of the program. That is.

  The third problem to be solved by the present invention is to prevent the performance of the application program from being seriously degraded due to the load of the added monitoring function as a result of adding a new monitoring function to the application program. is there.

  A representative invention disclosed in the present application is a computer system that executes an application program, wherein the computer system includes components of design information in one stage of the development process of the application program and a stage advanced from the one stage. The information indicating the correspondence relationship with the component of the design information is stored, and based on the information indicating the correspondence relationship, the component of the design information in one stage of the development process of the application program and the code of the application program Are associated with each other.

  According to an embodiment of the present invention, even if a business administrator or operation manager who does not understand the implementation of the program, the application program is monitored on the upstream application design information that can be understood from the business viewpoint. Can be specified. The program module for realizing the specified monitoring is immediately added or changed to the application program without requiring the work of the designer or programmer. In addition, it is possible to easily implement appropriate monitoring according to the execution state of the application program by cooperating a plurality of monitoring functions without requiring programming work. Furthermore, careless monitoring settings can be prevented from causing serious performance degradation in the execution of application programs, and application monitoring settings can be performed safely.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of an application program monitoring system according to the first embodiment of this invention.

  The known application server 160 reads the application program 170 from a known program storage device (not shown) and executes it. In the present embodiment, a case where the application server 160 is a Java 2 Enterprise Edition (J2EE) (Java is a registered trademark) application server and the application program 170 is an application program implemented using J2EE will be described as an example. In the present embodiment, the application server 160 includes a function for rewriting an application program 170 to be executed during execution using a bytecode instrumentation technique, and an interface (not shown) for controlling the rewriting function. .

  The user interface 110 is a user interface for setting and managing monitoring of application programs executed in this system.

  The correspondence search unit 115 uses the information stored in the design correspondence storage unit 130 to search the correspondence between the components of the design information created in the application development process stored in the design information storage unit 120. To do.

  Here, the design information is a general term for model diagrams and the like used when the application program 170 is developed. The components of the design information are points (locations) or data on the design information. For example, the component of the design information may be an activity or connection flow in a business process diagram, a message in a UML sequence diagram, or a class, method, or field in a class diagram. It may be a class, method or field in the program code.

  The management unit 141 generates a program module for executing monitoring of the application program 170 in the application server 160 according to the setting, and further inserts the generated program module into the application program 170. A program module that executes this monitoring is called a probe.

  The condition control unit 143 controls the monitoring operation of the probe based on a predetermined condition.

  The load control unit 145 measures and controls the influence on the application program 170 due to the load of the probe monitoring operation.

  A known monitoring event output unit 180 provides an application programming interface (API) that outputs a monitoring event to the application program 170. Examples of the monitoring event output unit 180 include Java Logging API or Log4J which is a Java logging API. The monitoring event output unit 180 outputs the monitoring event to a known monitoring event recording unit 190 in a format designated by the user. The monitoring event recording unit 190 may be a file system or a database, for example.

  FIG. 24 is a block diagram illustrating a physical configuration of the application program monitoring system according to the first embodiment of this invention.

  Specifically, FIG. 24 shows an example of a computer system that implements the system shown in FIG. This computer system includes a CPU 2703, a main memory 2705, an external storage device 2707, an input device 2709, an output device 2711, and a network interface 2713. These are connected by a bus 2701.

  The CPU 2703 is a processor that executes a program stored in the main memory 2705. In the following description, the processing executed by the program stored in the main memory 2705 is actually executed by the CPU 2703.

  The main memory 2705 is, for example, a semiconductor device. The main memory 2705 stores a software program read from the external storage device 2707, information referred to by the program, and the like.

  The input device 2709 is, for example, a keyboard or a mouse.

  The output device 2711 is, for example, a display device.

  The external storage device 2707 is, for example, a hard disk device. The design information storage unit 120, the design correspondence storage unit 130, and the monitoring event recording unit 190 illustrated in FIG. 1 are realized as storage areas in the external storage device 2707. These storage units may be implemented as a database, for example. Further, a software program constituting the application program monitoring system shown in FIG. 1 is stored in the external storage device 2707.

  Specifically, the software programs constituting the application program monitoring system are the user interface 110, the correspondence search unit 115, the management unit 141, the condition control unit 143, the load control unit 145, and the monitoring event output unit 180. These software programs are read into the main memory 2705 and executed by the CPU 2703, thereby realizing the system shown in FIG. The user interface 110 uses the input device 2709 and the output device 2711 to receive input from the user and output information to the user.

  When the system shown in FIG. 1 is realized by a single computer system, the external storage device 2707 further stores an application server 160 and an application program 170. In that case, the application server 160 and the application program 170 are read into the main memory 2705 and executed by the CPU 2703.

  The application program monitoring system shown in FIG. 1 can also be realized by a plurality of computer systems connected via a network interface 2713, instead of a single computer system. In that case, the units shown in FIG. 1 are separately arranged on a plurality of computer systems and connected via a network interface 2713.

  Next, a flow of processing in which the management unit 141 illustrated in FIG. 1 executes monitoring of an application program based on a given monitoring setting of the application program will be described with reference to FIGS. 3 and 4.

  FIG. 3 is an explanatory diagram illustrating a configuration of the management unit 141 according to the first embodiment of this invention.

  The management unit 141 includes a probe definition storage unit 1601, a probe generation unit 1604, and a probe insertion unit 1608. The probe definition storage unit 1601 is a storage area secured in the main memory 2705 or the external storage device 2707. The probe generation unit 1604 and the probe insertion unit 1608 are program modules included in the management unit 141. The user interface 110, application server 160, application program 170, monitoring event output unit 180, and monitoring event recording unit 190 are the same as those shown in FIG.

  The application program 170 is read into the application server 160 and executed.

  The probe 1610 is a program module that is inserted into the application program 170 and executes monitoring of the application program 170.

  The probe definition storage unit 1601 stores a probe definition that is a monitoring setting of the application program 170 described at the program code level. According to the probe definition, the “location (point)” for executing monitoring in the application program and the “item” to be monitored are explicitly specified using components of the program code.

  FIG. 4 is an explanatory diagram illustrating a configuration example of the probe definition storage unit 1601 according to the first embodiment of this invention.

  The probe definition stored in the probe definition storage unit 1601 includes a probe ID 1001, a probe number 1002, a probe name 1003, a probe insertion point 1012, and monitoring data 1021.

  The probe ID 1001 is an ID for uniquely identifying the probe 1610, and is used for identifying the probe 1610 and the monitoring data output by the probe 1610.

  The probe number 1002 is a serial number for identifying a plurality of probe definitions having the same probe ID 1001 when they exist. In this embodiment, since there is only one probe definition having one probe ID, the probe number 1002 is always “1”. Therefore, the probe number 1002 is not an essential item in the present embodiment.

  The probe name 1003 is a name given to the probe 1610.

  The probe insertion point 1012 designates a place (point) where the probe 1610 is inserted. Specifically, the probe insertion point 1012 is specified by a component of the application program.

  The monitoring data 1021 specifies data to be acquired and output by the probe 1610. The data acquired and output by the probe 1610 is normally acquired as a so-called application log. The probe 1610 acquires and outputs data such as input / output parameters processed by the application program 170 or processing time of the application program 170 as an application log (see S305 in FIG. 3).

  In the example of FIG. 4, in the probe definition on the first line, the probe 1610 whose probe ID 1001 is “P001” is inserted into the method “FuncA.method1 (String d1, String d2)”, and the first argument “d1” of the method "Is acquired as monitoring data. Further, “_responseTime” specified in the item of the monitoring data 1021 on the third line indicates the processing time of the method. That is, the probe definition in the third row indicates that the probe 1610 with the probe ID 1001 “P003” is inserted into the method “FuncC.method3 (Data2 d)” and the processing time of the method is acquired as monitoring data. ing.

  In the present embodiment, the probe 1610 is realized as advice in aspect-oriented programming, which is a known technique. The probe insertion unit 1608 inserts the probe 1610 into the application program 170 by incorporating the program code of the probe 1610 defined as advice into the application program 170 by the aspect-oriented programming framework. The point at which the probe 1610 is inserted is specified by using a point cut created based on the probe insertion point 1012 specified in the probe definition as an input to the aspect-oriented programming framework.

  The probe 1610 is inserted into the application program 170 dynamically during the execution of the application program 170 by using an aspect-oriented programming framework (Dynamic AOP framework) that supports the incorporation of dynamic advice. Can do. An example of a Dynamic AOP framework is JBoss AOP.

  Note that by using any technique for dynamically updating the program code during the execution of the application program 170, the form of the probe 1610 and the insertion of the probe 1610 into the application program 170 can be realized.

  The probe 1610 may be inserted when the application program 170 is not executed. However, if the probe insertion unit 1608 does not have a function of inserting the probe 1610 during the execution of the application program 170, when the probe 1610 is inserted or removed, all or part of the application program 170 is stopped and then restarted. There is a need to.

  Next, a flow of processing for executing monitoring of the application program 170 based on the probe definition will be described with reference to FIG. First, the probe generation unit in the management unit 141 acquires a probe definition from the probe definition storage unit 1601 in response to an external request (S301) (S302). This request is, for example, a request input by the business manager or the operations manager using the user interface 110. Then, the probe generation unit 1604 acquires the designated data, and generates the program code of the probe 1610 that outputs the data as monitoring data as the advice of aspect-oriented programming based on the contents of the monitoring data 1021 in the probe definition. To do.

  A known program code automatic generation technique is used for code generation of the probe 1610. Specifically, the probe generator 1604 given the probe insertion point 1012 and the monitoring data 1021 as inputs automatically generates a program code for the probe 1610 according to the inputs. The program code generated at this time is the point on the code of the application program 170 corresponding to the input probe insertion point 1012, acquires data corresponding to the input monitoring data 1021, and the data is sent to the monitoring event output unit The program code is output using 180 APIs (S305, S306).

  FIG. 5 is an explanatory diagram illustrating a code example of the probe 1610 generated as advice in the first exemplary embodiment of the present invention.

  FIG. 5 shows an example of the probe 1610 generated based on the probe definition whose probe ID 1001 exemplified in FIG. 4 is “P002”. The probe 1610 is generated by the probe generation unit 1604 as one class that inherits the Interceptor class. One probe advice is generated for one probe. The Static field logger (5th line) is an interface to the monitoring event output unit 180 that outputs monitoring data, and is set by an initialization code (not shown) executed when the probe 1610 is inserted.

  The probe advice method “invoke” is code for executing application program monitoring. The code for the invoke method is inserted at a specified point in the application program 170 by the Dynamic AOP framework.

  In the invoke method, first, using the API provided by JBoss AOP, the object of the first argument of the method in which the probe is inserted is acquired (line 8). Furthermore, the value of a specific field is acquired from the object (line 9), and the monitoring data specified by the probe definition is acquired. Then, the monitoring event object event to be output is generated (line 10). A probe ID is set in the monitoring event object. This probe ID is used to identify monitoring data. Next, the generated event object is output to the monitoring event output unit 180 using the logger interface (11th line). Thereafter, processing of the method main body in which the probe 1610 is inserted is executed (line 12).

  The probe insertion unit 1608 dynamically inserts the probe 1610 generated by the probe generation unit 1604 into the application program 170 using the Dynamic AOP framework (see S304 in FIG. 3). A point cut that specifies the insertion position of the probe 1610 is generated based on the probe insertion point 1012 of the probe definition. For example, in the probe definition shown in FIG. 4, the pointcut that specifies the insertion position of the probe 1610 with the probe ID “P002” is “execution (* FuncB-> method2 (Data1))”.

  In order to insert the probe 1610, the probe insertion unit 1608 gives the generated probe aspect class object, point cut, and probe ID to the aspect manager provided by JBoss AOP (see FIG. 6).

  FIG. 6 is an explanatory diagram illustrating an example of a program code for adding the probe 1610 according to the first embodiment of this invention to the application program 170.

  FIG. 6 shows, as an example, program code generated by the probe insertion unit 1608 for inserting the probe 1610 shown in FIG. 5 into the application program 170. Hereinafter, a procedure of processing executed by the program code of FIG. 6 will be described.

  First, an AdviseBinding object binding of a point cut that specifies a position where the probe 1610 is inserted is generated (first line).

  Next, a probe ID is set as the name of the generated AdviseBinding object (second line), and a probe advice class is set (third line).

  Next, this AdviseBinding object is registered in the aspect manager (line 4).

  The aspect manager inserts the program code of the probe advice by rewriting the byte code of the application program 170 according to the pointcut and advice class set in the registered AdviseBinding object.

  The probe insertion unit 1608 can also remove the probe 1610 already inserted into the application program 170 from the application program 170 using the function of the Dynamic AOP framework in response to a request from the outside. The removal of the probe 1610 is executed by calling the removeBinding method of the aspect manager with the ID of the probe 1610 to be removed as an argument. The aspect manager rewrites the byte code of the application program 170 to remove the program code of the advice set in the AdviseBinding object whose name is the probe ID passed as an argument.

  Next, the correspondence search unit 115 uses the design correspondence relationship to search for a point on the program code corresponding to the designation on the design information of “point to execute monitoring”, that is, a probe insertion point in the probe definition. The flow of processing will be described.

  First, the design correspondence will be described with reference to the example of FIG.

  FIG. 7 is an explanatory diagram illustrating an example of the design correspondence relationship according to the first embodiment of this invention.

  The design correspondence relationship is stored in the design correspondence relationship storage unit 130. The design correspondence includes a correspondence source element 601 and a correspondence destination element 603. The correspondence element 601 stores design information components created at any stage of the application program development process. The correspondence destination element 603 stores a component having a design correspondence relationship with each component stored as the correspondence source element 601. The component corresponding to the design correspondence is a component of the design information created at a more advanced stage of the development process. For example, the correspondence source element 601 “A01.function 1” and the correspondence destination element 603 “B02.function 1.processing 1 (data 1 X)” are stored as the design correspondence 611 in FIG. This means that the component “A01.Function 1” in the development process at a certain stage becomes the component “B02.Function 1.Process 1 (Data 1 X)” in the development process advanced by one stage. .

  Note that one or a plurality of components may be in a design correspondence relationship with respect to one component. In the present embodiment, a case will be described in which only one component corresponds to one component. The case where a plurality of components correspond to one component will be described later in the second embodiment.

  In order to identify an arbitrary constituent element over a plurality of design information and express a design correspondence between the constituent elements, each constituent element of the design information is given an ID for uniquely identifying them. This ID is stored as design correspondence information. In this embodiment, this ID is called an object ID.

  In the present embodiment, the object ID is composed of an ID of design information including a component identified by the object ID and an ID of each component. For example, in the example of FIG. 7, the object ID “A01.function 1” ”of the corresponding element of the design correspondence 611 indicates an activity element representing the function 1 that exists in the business process diagram with the ID A01. The information ID is represented by an alphabet (for example, A) indicating the stage of the development process of the application program 170 and a serial number (for example, 01) of design information at each stage.

  The object ID only needs to uniquely identify the component of the design information. Further, it is only necessary that design information including the component pointed to by the object ID can be specified. Therefore, it is desirable that the object ID includes the design information ID as a part of the object ID.

  FIG. 7 shows the design correspondence relationship between the two activity elements (function 1 and function 2) of the business process diagram of stage A for the application program 170 developed using the design information of the three stages of A, B and C. Show. Specifically, the design correspondence information in the example of FIG. 7 includes the design correspondence relationship (611, 612, 614, 617, 631, 632) of the processing (method), and the design correspondence relationship of the input data of the processing (615, 616). And the design correspondence (620 to 622) of the data structure used for the input data of the process.

  In order to show the design correspondence of the input data of the process, the object ID of each input data is used. The object ID of the input data is configured by concatenating the process name and the input data expression. For example, in the design correspondence relationship 615, the input data “Xa” of the method “Function 1. Processing 1” in the design information B02 and the input data d1 of the method “FuncA.method1” in the design information C 02 correspond to each other. Is shown.

  The design correspondences 631 and 632 indicate that the constituent elements of the design information C02 correspond to specific methods of the installed application program 170.

  In the present invention, generation of design correspondence information and accumulation of the design correspondence information in the design correspondence information storage unit 130 can be performed by an arbitrary method. However, appropriate design correspondence information needs to be stored for each design information created in the development process of the application program 170.

  The generation of the design correspondence information and the accumulation in the design correspondence information storage unit are preferably executed every time design information that has advanced in the development process is created in the development process of the application program 170. Therefore, it is desirable that an application design tool used in the development process of the application program 170 has a function of generating and storing design correspondence information.

  Next, the program code using the stored design correspondence relationship is inserted from the “monitoring point (that is, the point to be monitored)” designated on the design information to the point where the designated monitoring probe is to be inserted. The flow of processing for searching by level will be described with reference to FIG.

  FIG. 8 is a flowchart of processing for searching for a point at which the probe 1610 is inserted by the correspondence search unit 115 according to the first embodiment of this invention.

  In the description of FIG. 8, the example of the design correspondence shown in FIG. 7 is used.

  A point to be monitored is specified by a user (for example, a business administrator or an operation manager) specifying a component of design information. The object ID of the designated component is input to the correspondence search unit 115.

  First, in step 1201, the correspondence search unit 115 initializes the variable oid with the input object ID.

  Next, in step 1202, the correspondence search unit 115 accesses the design correspondence storage unit 130 using oid as a key, and searches for a design correspondence having oid as the correspondence source element 601. Then, the correspondence search unit 115 stores the object ID stored in the correspondence destination element 603 of the design correspondence acquired as a result of this search in the variable nextOid.

  In step 1203, the correspondence search unit 115 determines whether nextOid is a special object ID indicating that it corresponds to a component of the implementation program code of the application program 170. The component of the implementation program code is, for example, a specific method.

  If it is determined in step 1203 that nextOid is a special ID for identifying a component of the implementation program code of the application program 170, the retrieved component corresponds to a specific method or the like. In this case, the process proceeds to step 1205, and the correspondence search unit 115 outputs the program code element indicated by nextOid as a search result, and ends the process.

  On the other hand, if it is determined in step 1203 that nextOid is not a special ID for identifying the component of the implementation program code of the application program 170, the component of the implementation program code corresponding to the designated component has not yet been searched. Absent. In this case, in step 1204, the correspondence search unit 115 updates the oid with the searched nextOid value, and returns to step 1202.

  By repeating the processing of step 1202 to step 1204, the design information components at each stage of the development process are searched step by step from the design information components at the initial stage of the development process having a high degree of abstraction. The program code component is retrieved.

  For example, the case where the activity “function 1” of the business process diagram “A01” is designated as a monitoring point in the example of the design correspondence relationship shown in FIG. 7 will be described. In this case, the object ID “A01.function 1” becomes the initial value of oid (step 1201). Thereafter, the correspondence search unit 115 sequentially searches for design correspondences by repeatedly executing Steps 1202 to 1204.

  As a result of the search, “B02.Function 1.Processing 1 (Data 1 X)” corresponding to “A01.Function 1” is acquired (corresponding relationship 611), and “B02.Function 1.Processing 1 (Data 1 X)” is obtained. “C02.FuncA.method1 (String d1, String d2)” corresponding to “C02.FuncA.method1 (String d1, String d2)” and “CODE.FuncA.method1 (corresponding relationship 614)” are acquired. String d1, String d2) ”is acquired (correspondence 631).

  Since “CODE.FuncA.method1 (String d1, String d2)” is a component of the program code (step 1203), the method “FuncA.method1 (String d1, String d2)” is the program code corresponding to function 1. The element is searched for (step 1205). By inserting the probe 1610 into this method, monitoring in “Function 1” is executed. If the target to be monitored at the specified point does not need to generate the probe to be inserted individually, it is specified by inserting probe advice prepared in advance at the point on the searched program code. Execute monitoring. The monitoring target that does not need to individually generate the probe to be inserted is, for example, processing time.

  Through the above processing, a point on the program code corresponding to the “point to be monitored” designated on the design information is retrieved using the stored design correspondence. Then, the searched points are monitored. As a result, it is possible to specify monitoring of the application program 170 on the design information created in the application program development process without having to be aware of the installed program code.

  FIG. 8 shows the processing of the correspondence search unit 115 when “monitoring point” is designated on the design information. On the other hand, arbitrary data on the design information may be designated as “data to be monitored (that is, data to be monitored)”. Arbitrary data on design information is data processed in a business process, and is, for example, an input / output parameter of any process. Next, the flow of processing executed by the correspondence search unit 115 when “data to be monitored” is designated on the design information will be described. In this case, the correspondence search unit 115 searches the component of the program code corresponding to the designated “data to be monitored”, that is, the monitoring data 1021 in the probe definition with reference to the design correspondence.

  The search for “data to be monitored” is executed by the same process as the search for “point to be monitored” described above.

  The design correspondence includes not only the “processing” correspondence between the design information but also the “data structure” and “input / output parameters”. For example, in FIG. 7, design correspondences 615 and 616 indicate correspondences between the input parameters of the processes having the design correspondence. The design correspondence 615 indicates that the field “a” of the input parameter “X” of the process 1 corresponds to the first argument of the method “method1” corresponding to the process 1.

  The design correspondences 620 to 622 indicate the correspondence between the data structure “data 1” and each field of the data structure “Data1”.

  For example, in the class diagram “B02” at a certain stage of the application program development process, when specifying to monitor the field “b” for the input parameter “X” of the method “Process 1” of the class “Function 1”, The operation manager or the business manager gives the object ID “B02. Function 1. process 1 # Xb” as an input to the correspondence search unit. Thereafter, by following the design correspondence according to the flow shown in FIG. 8, in the implemented program, the second argument “d2” in the method “method1” of the class “FuncA” corresponds to the specified monitored data. Is being searched. In the case of this example, the monitoring point is also designated at the same time as the monitoring data. Therefore, the search determines both the point where the probe is inserted and the data acquired by the probe, and the designated monitoring is executed.

  The design correspondence relationship storage unit may store design correspondence relationships for all components that can be designated as points to be monitored and data. However, when it is possible to infer another design correspondence by combining a plurality of design correspondences or information on design information, it is possible to store the design correspondence by omitting the storage of the design correspondence that can be inferred. The cost required for this can be reduced.

  The example of the design correspondence 617 in FIG. 7 indicates that the input parameter “X” in the processing of the correspondence source corresponds to the input parameter “d1” of the processing of the correspondence destination. However, the design correspondence relationship for each field of the input parameter “X” is not stored. However, it is stored by using the design correspondences 620 to 622 of “B11.Data1” which is the data structure of the input parameter “X” and “C11.Data1” which is the data structure of the input parameter “d1”. It can be inferred that there is no design correspondence. The processing flow will be described with reference to FIG.

  FIG. 9 is a flowchart of processing in which the correspondence search unit 115 according to the first embodiment of this invention infers the design correspondence.

  This process corresponds to the process of step 1202 in the flow for specifying a program code element using the design correspondence shown in FIG.

  First, in step 1301, the correspondence search unit 115 stores the input object ID in a variable oid. In the above example, “B02.Function 2.Process 2 # X.a”, that is, the field “a” of the parameter “X” of “Process 2” is input as the object ID.

  Next, in step 1303, the correspondence search unit 115 accesses the design correspondence storage unit 130 using oid as a key, and searches for a design correspondence having oid as the correspondence source element 601.

  Next, in step 1305, the correspondence search unit 115 determines whether there is a design correspondence relationship regarding oid. Specifically, it is determined whether or not the design correspondence relationship in which the correspondence source element 601 is oid is stored in the design correspondence relationship storage unit 130.

  If it is determined in step 1305 that there is a design correspondence related to oid, the process of the correspondence search unit 115 proceeds to step 1307.

  In step 1307, the correspondence search unit 115 outputs the searched correspondence destination element 603 of the design correspondence as a search result, and ends the processing.

  If it is determined in step 1305 that there is no design correspondence related to oid, the process of the correspondence search unit 115 proceeds to step 1309. For example, when the oid is “B02.Function 2.Process 2 # X.a”, there is no design correspondence having the oid as the correspondence source element 601. In this case, the process proceeds to Step 1309.

  In step 1309, the correspondence search unit 115 identifies a constituent element that is higher than the constituent element indicated by oid. For example, when oid is “B02.Function 2.Processing 2 # X.a”, “B02.Function 2.Processing 2 # X” with the field “a” deleted is a higher-level component. In this case, the correspondence search unit 115 searches for the design correspondence using “B02.function 2.processing 2 # X.a” as a new oid instead of “B02.function 2.processing 2 # X.a”.

  Next, in step 1311, the correspondence search unit 115 determines whether or not a design correspondence relationship related to oid (that is, a higher-order component specified in step 1309) is stored. This determination is performed in the same manner as in step 1305.

  If it is determined in step 1311 that the design correspondence related to oid is not stored, the correspondence retrieval unit 115 returns to step 1309 to identify and retrieve a higher-order component.

  If it is determined in step 1311 that the design correspondence relationship regarding oid is stored, the correspondence search unit 115 proceeds to step 1313.

  In step 1313, the correspondence search unit 115 acquires a component having a design correspondence relationship with a higher-order component. For example, when the upper component is “B02.Function 2.Process 2 # X”, this component is the process “B02.Function 2.Process 2” having one parameter X. This corresponds to B02.Function 2.Processing 2 (data 1 X) which is a corresponding element of the design correspondence 617, and the design correspondence 617 related to this is stored. In this case, “C02.FuncB.method2 # d1” is acquired as a component corresponding to the upper component.

  In step 1315, the correspondence search unit 115 estimates the design correspondence of the lower component based on the design correspondence of the higher component searched in step 1313. For example, when “C02.FuncB.method2 # d1” is acquired in step 1313, the correspondence search unit 115 sets the design correspondence of the field “a” for the data structure “B11.data 1” of the parameter “X”. Search for. As a result, it is searched that “fieldA” in the data structure “C11.Data1” of the parameter “D1” corresponds (see design correspondence 621). Therefore, the correspondence search unit 115 infers “C02.FuncB.method2 # d1.fieldA” as a component having a design correspondence with “B02.Function 2.Process 2 #Xa”, and the design correspondence search result. Output as.

  The identification of the upper component in the above flow and the analogy of the design correspondence based on it are executed according to the rules defined in advance according to the design information in which the component exists and the type of the component. According to this rule, a part of the design correspondence information stored in the design correspondence storage unit 130 can be omitted, and the storage cost can be reduced.

  Next, the user interface 110 shown in FIG. 1 will be described.

  FIG. 10 is an explanatory diagram of an example of the user interface 110 that executes the monitoring setting of the application program 170 on the design information according to the first embodiment of this invention.

  The user interface 110 shown in FIG. 10 is a point where a user (for example, an operation manager or a business manager) executes monitoring in the application program 170 based on design information created at one design stage of application program development ( Monitoring point) and data to be monitored (monitoring data) can be input.

  The design information display window 550 is a window for displaying design information for inputting contents to be monitored by the user. The user designates a monitoring point by selecting the component 551 of the design information 552 displayed in this window, for example, by clicking the component.

  When a monitoring point is specified, the user interface 110 displays a list 562 of items that can be monitored at the specified monitoring point in the monitoring data selection window 560. The user can select data to be monitored at the designated monitoring point by checking the radio button 561 corresponding to each data item displayed in the item list 562.

  After selecting the monitoring data, when the user operates the determination button 563, the specified monitoring point and monitoring data are determined, and the search by the correspondence search unit 115 is executed.

  When data that is not unique to a monitoring point, such as processing time, is specified as the monitoring data, the search by the correspondence search unit 115 is executed using the object ID of the specified monitoring point as a key (see FIG. 8). . The components of the searched program code are stored in the probe insertion point 1012 of the generated probe definition. The probe definition monitoring data 1021 stores a predefined symbol representing the selected monitoring data such as the processing time.

  When the designated monitoring data is an input / output parameter or the like unique to the monitoring point, a search by the correspondence search unit 115 is executed using the object ID representing the data as a key (see FIG. 8). As a result, the constituent element of the program code to be searched has information on both the point on the program code corresponding to the designated monitoring point and the data handled at that point. Therefore, according to the searched component, a point on the program code corresponding to the monitoring point is stored in the probe insertion point 1012 of the generated probe definition, and data handled at the point is stored in the monitoring data 1021. The

  When storing the probe definition, the user interface 110 displays an interface that allows the user to input a name to be given to the input monitoring setting. The user interface 110 stores the input name in the probe name 1003 of the probe definition. Further, the user interface 110 generates an ID for uniquely identifying the probe definition, and stores the ID as the probe definition ID 1001.

  FIG. 11 is an explanatory diagram illustrating an example of a monitoring setting table according to the first embodiment of this invention.

  The monitoring setting table is a table for managing the monitoring setting input by the user and the probe 1610 generated based on the monitoring setting. The monitoring setting table may be a part of the user interface 110 or a part of any storage unit (for example, the design information storage unit 120) that can be accessed by the user interface 110. In any case, the monitoring setting table is stored in the main memory 2705 or the external storage device 2707.

  ID 901 stores an ID for uniquely identifying each monitoring setting.

  The monitoring setting name 903 stores the name assigned to the monitoring setting.

  The monitoring point 912 stores the object ID of the input location to be monitored.

  The monitoring data 921 stores an object ID of the input monitoring data or a symbol of a monitoring data item defined in advance. The symbol of the monitoring data item defined in advance is, for example, “__responseTime”.

  The ID 901 and the monitoring setting name 903 of the monitoring setting table store the same as the ID 1001 and probe name 1003 of the generated probe definition. By using the same ID in the monitoring setting and the probe definition, the monitoring setting and the probe definition are associated with each other.

  The user interface 110 refers to the information of the monitoring point 912 and the monitoring data 921 stored in the monitoring setting table, displays the information of the set probe 1610 on the design information, and further allows the user to select it. .

  For example, when the user operates the screen shown in FIG. 10 and specifies that the data corresponding to the monitoring data “data 1.a” is to be monitored at the location corresponding to the monitoring point “A01. Function 1”. “A01.Function 1” is stored in the monitoring point 912 of the monitoring setting table, and “Data 1.a” is stored in the monitoring data 921. Further, information for identifying the monitoring point or the like is stored in the ID 901 and the monitoring setting name 903. In the example of FIG. 11, “P001” and “processing 1 input data” are stored as the ID 901 and the monitoring setting name 903 for identifying the monitoring point “A01.function 1” and the monitoring data “data 1.a”, respectively. (See the first line in FIG. 11).

  In this case, the correspondence search unit 115 receives “A01.function 1” and “data 1.a” as inputs and executes the processes shown in FIGS. As a result, for example, “FuncA.method1 (String d1, String d2)” and “d1” are acquired as the probe insertion point and the monitoring data, respectively. In this case, “P001” and “process 1 input data” are stored as the ID 1001 and the probe name 1003 in the probe definition storage unit 1601 as in the monitoring setting table. Then, “FuncA.method1 (String d1, String d2)” and “d1” acquired as a result of the search are stored as the probe insertion point 1012 and the monitoring data 1021 (see the first line in FIG. 4).

  As design information displayed in the design information display window 550 for designating a monitoring point and monitoring data, any design information created at any stage of the application development process according to a user input can be used. At this time, the design information including the components having the design correspondence with the component of the displayed design information is searched using the design correspondence information, and the searched design information is displayed as a candidate of the design information to be displayed. The user can also select one of the displayed candidates.

  Further, the user interface 110 is provided with a user interface for inputting the value of the monitoring definition ID to be assigned to the monitoring definition to be set and the value of the monitoring item ID to be assigned to each monitoring item set in the monitoring definition in the setting of the monitoring definition. Also good. Alternatively, the user interface 110 may automatically generate a monitoring definition ID and a monitoring item ID and set them.

  The configuration and input procedure of the user interface 110 for the monitoring setting are not limited to the example shown in the present embodiment, and the monitoring point, the item monitored at the monitoring point, and the monitoring definition ID as necessary. And any interface that can input the monitoring item ID.

  As described above, using the system shown in FIG. 1, when the user sets monitoring of an application program to be executed in the design information at one stage of the application program development process, the setting between the input setting and the implemented program is set. The association is automatically executed without human intervention. Therefore, it is possible to immediately monitor the set application program.

  Next, the configuration and operation of a system that controls the monitoring operation of the probe 1610 inserted into the application program 170 in accordance with specified conditions will be described.

  FIG. 12 is a block diagram illustrating a configuration of a system that controls the monitoring operation of the probe 1610 according to the first embodiment of this invention in accordance with conditions.

  The condition control unit 143 includes a probe control definition storage unit 1602, a setting unit 1609, a condition determination unit 1606, and a probe control unit 1607. The management unit 141 is the same as that shown in FIG.

  The probe control definition storage unit 1602 is a storage area secured in the main memory 2705 or the external storage device 2707. The condition determination unit 1606, the probe control unit 1607, and the setting unit 1609 are program modules included in the condition control unit 143.

  The probe 1610 is a program module inserted into the application program 170 by the management unit 141 shown in FIG. One or more probes 1610 may be inserted into the application program 170.

  The monitoring event output unit 180a is obtained by adding a collector 1620 to the known monitoring event output unit 180 shown in FIG.

  The collector 1620 is a program module that acquires data output by the probe 1610 and transfers the data to the condition determination unit 1606 as necessary.

  The data acquisition unit 1630 is a program module that is arranged in the application server 160 or the like and acquires state monitoring data and external information. FIG. 12 shows an example in which a data acquisition unit 1630 is arranged on the application server 160.

  The data acquisition unit 1630 acquires information other than the data monitored by the probe 1610. Specifically, the data acquisition unit 1630 acquires, for example, the application program 170, the application server 160, the OS and hardware that executes the application server 160, or the state monitoring data of the system itself that monitors the application program. For example, the data acquisition unit 1630 may acquire the load of the CPU 2703 as the state monitoring data.

  The data acquisition unit 1630 may acquire external information such as time in addition to the state monitoring data.

  The probe control definition storage unit 1602 stores the probe control definition. The probe control definition is information that specifies a condition for controlling the monitoring operation of the probe 1610 inserted in the application program 170 and an operation for controlling the monitoring operation of the probe 1610. The monitoring operation of the probe 1610 is, for example, acquisition or output of monitoring data.

  Based on this probe control definition, the setting unit 1609 sets each unit and controls the monitoring operation of the probe 1610.

  First, the flow of processing for controlling the monitoring operation of the probe 1610 based on the probe control definition will be described.

  FIG. 13 is an explanatory diagram illustrating a configuration example of the probe control definition storage unit 1602 according to the first embodiment of this invention.

  The probe control definition stored in the probe control definition storage unit 1602 includes an ID 1801, a control condition 1803, and a control operation 1805. ID 1801 identifies each probe control definition. The control condition 1803 specifies a condition for controlling the monitoring operation of the probe 1610. A control operation 1805 specifies an operation for controlling the monitoring operation of the probe 1610.

  Control of the monitoring operation of the probe 1610 is described in IF-THEN format. Specifically, the IF portion corresponds to the control condition 1803 and the THEN portion corresponds to the control operation 1805. In other words, when the condition stored in the control condition 1803 is satisfied, the operation stored in the control operation 1805 is executed.

  The control condition 1803 is described by a conditional expression including data used for condition determination. Monitoring data acquired by the probe 1610 inserted into the application program 170 can be designated as data used for condition determination. Alternatively, information acquired by the data acquisition unit 1630 prepared in advance in the system may be designated.

  Data used for the condition determination is referred to as determination information. Each determination information is identified by a unique ID. The conditional expression representing the control condition is described using the ID of the determination information. When the monitoring data acquired by the probe 1610 is used as determination information, the probe ID 1001 in the probe definition illustrated in FIG. 4 is the ID of the determination information. Data acquired by the data acquisition unit 1630 is specified by a special ID assigned in advance.

  The control operation 1805 stores a control method executed when the control condition 1803 is satisfied, and a probe 1610 whose monitoring operation is controlled by the method. Further, if necessary, control parameters are also stored.

  The control method of the monitoring operation of the probe 1610 is, for example, the start or stop of the monitoring operation by the probe 1610. Alternatively, the monitoring operation of the probe 1610 may be controlled by executing acquisition and output of monitoring data at predetermined intervals set by time or the number of executions of the application program 170.

  The start or stop of the monitoring operation of the probe 1610 can be executed by the probe insertion unit 1608 inserting the probe 1610 into the application program 170 (remove) or removing the probe from the application program 170 (remove). . Alternatively, an operation control code (not shown) for controlling the execution of the monitoring operation may be incorporated in the probe 1610 and a variable for controlling the operation control code may be assigned. This variable is called an operation control variable. The monitoring operation of the probe can be controlled (dissable or started) by determining whether or not the monitoring operation can be executed by referring to the operation control variable. The operation control variable is stored in a memory area accessible from both the probe 1610 and the probe control unit 1607.

  The first line in FIG. 13 indicates that the probe with the probe ID P004 is removed from the application program 170 when the average value of the monitoring data of the probe 1610 with the probe ID P003 exceeds 10 for a certain period. ing. The second line indicates that when the value of the monitoring data of the probe P002 is less than 5, the operation of the probe P005 is stopped. The third line indicates that the operation of the probe P006 is started when the value of the probe P001 is a character string starting with A.

  A flow of processing for controlling the monitoring operation of the probe 1610 by the above system will be described.

  First, in accordance with the execution of the application program 170, the probe 1610 inserted into the application program 170 outputs monitoring data (S1201). The output monitoring data is output via the monitoring event output unit 180a. At this time, the collector 1620 in the monitoring event output unit 180a determines whether the monitoring data is data used as determination information. If it is determined that the monitoring data is data used as determination information, the monitoring event output unit 180a transfers the monitoring data to the condition determination unit 1606 (S1202).

  On the other hand, when the determination information is data other than the monitoring data of the probe 1610, the data acquisition unit 1630 periodically acquires the data and transmits the acquired data to the condition determination unit 1606 (S1203). Alternatively, data used as determination information may be acquired by the condition determination unit 1606 calling the data acquisition unit 1630 periodically.

  The condition determination unit 1606 evaluates the conditional expression stored in the control condition 1803 using the determination information acquired from the collector 1620 or the data acquisition unit 1630, and transmits the evaluation result to the probe control unit 1607 (S1204). The probe control unit 1607 controls the monitoring operation of the probe 1610 by executing the operation stored in the control operation 1805 according to the condition determination result received from the condition determination unit 1606 (S1205).

  Next, refer to FIGS. 14 and 15 for the flow of processing for setting the condition determination unit 1606 and the probe control unit 1607 so that the setting unit 1609 controls the monitoring operation of the probe 1610 based on the probe control definition. To explain.

  FIG. 14 is a flowchart of processing in which the setting unit 1609 according to the first embodiment of this invention sets the condition determination unit 1606.

  The setting unit 1609 sets the condition determination unit 1606 by executing the flow of FIG. 14 based on the probe definition.

  First, in step 2001, the setting unit 1609 acquires a probe control definition from the probe control definition storage unit 1602.

  Next, in step 2003, the setting unit 1609 analyzes the probe definition control condition 1803 and extracts designation of determination information.

  Next, the setting unit 1609 executes the processes in steps 2005 to 2008 for all the determination information extracted in step 2003.

  In step 2005, the setting unit 1609 determines whether the data is data acquired by the probe 1610, based on the data ID specified as the determination information. If it is determined in step 2005 that the specified data is data acquired by the probe 1610, the process proceeds to step 2006. On the other hand, if it is determined in step 2005 that the designated data is not data acquired by the probe 1610, the process proceeds to step 2007.

  In step 2006, the setting unit 1609 registers the ID of the designated data in the collector 1620 of the monitoring event output unit 180a.

  In step 2007, the setting unit 1609 specifies the data acquisition unit 1630 that acquires data based on the ID of the data specified as the determination information, and sets the specified data acquisition unit 1630 to acquire data. .

  After step 2006 or 2007 is executed, the setting unit 1609 determines in step 2008 whether or not the processing of all the determination information extracted in step 2003 has been completed.

  If it is determined in step 2008 that all the determination information processes have been completed, the setting unit 1609 ends the process. On the other hand, if it is determined in step 2008 that all the determination information has not been processed, there is unprocessed determination information. In this case, the setting unit 1609 returns to step 2005 in order to process unprocessed determination information.

  The collector 1620 determines whether or not the ID added to the monitoring data is the ID registered in step 2006 when the probe 1610 acquires the monitoring data and outputs the monitoring data via the monitoring event output unit 180a. When it is determined that the ID is registered, the collector 1620 transmits the monitoring data to the condition determination unit 1606.

  The condition determination unit 1606 interprets and executes the condition determination expression described as the control condition 1803 using the value of the determination information. Specifically, a known interpreter included in the condition determination unit 1606 may interpret the condition determination expression and execute it. Alternatively, a program module that executes a condition determination expression may be generated and managed in advance by a known compiler, and the program module may interpret and execute the condition determination expression.

  In place of the condition determination expression, a condition determination program code described in a program language that can be interpreted and executed by the condition determination unit 1606 may be stored as the control condition 1803. In that case, the condition determination unit 1606 calls the condition determination program code with the determination information necessary for the condition determination as an input. The condition determination program code executes the condition determination by returning the condition determination result.

  FIG. 15 is a flowchart of processing for setting the probe control unit 1607 by the setting unit 1609 according to the first embodiment of this invention.

  The setting unit 1609 sets the probe control unit 1607 by executing the flow shown in FIG. 15 based on the probe definition.

  In step 2101, the setting unit 1609 acquires a probe control definition from the probe control definition storage unit 1602.

  Next, in step 2102, the setting unit 1609 analyzes the probe definition control operation to identify the ID of the probe 1610 to be controlled and the operation to control the monitoring operation of the probe 1610.

  Next, in step 2104, the setting unit 1609 is designated such that the monitoring operation control of the probe 1610 is executed by inserting the probe 1610 into the application program 170 or removing the probe 1610 from the application program. It is determined whether or not. If it is determined in step 2104 that the control is executed by inserting or removing the probe 1610, the process is terminated. If it is determined that control is not performed by insertion or removal of the probe 1610, the process proceeds to step 2105.

  In step 2105, the setting unit 1609 causes the management unit 141 to generate a new probe 1610 in which an operation control code for controlling the monitoring operation of the probe 1610 is incorporated in the probe 1610 specified in step 2102. The management unit 141 regenerates the program code of the specified probe 1610 based on the probe definition. At that time, the management unit 141 incorporates the operation control code into the regenerated probe 1610.

  In step 2106, the management unit 141 inserts the probe 1610 incorporating the operation control code generated in step 2105 into the application program 170. If the probe 1610 identified in step 2102 has already been inserted, the program code of the old probe 1610 is replaced with the program code of the probe generated in step 2105.

  With the above processing, functions necessary for controlling the monitoring operation of the probe 1610 are set.

  FIG. 16 is a flowchart of processing in which the probe control unit 1607 according to the first embodiment of this invention controls the operation of the probe 1610.

  In step 1901, the probe control unit 1607 determines whether to control the monitoring operation of the probe 1610 by inserting or removing the probe 1610. For this determination, for example, the probe control unit 1607 may be provided with a table (not shown) using the probe control definition ID 1801 as a key. The table stores a method for controlling the monitoring operation of the probe 1610. The probe control unit 1607 can execute the determination in step 1901 by referring to this table.

  If it is determined in step 1901 that the monitoring operation control of the probe 1610 is executed by inserting or removing the probe 1610, the process proceeds to step 1902. On the other hand, if it is determined that the monitoring operation control of the probe 1610 is not executed by insertion or removal of the probe 1610, the process proceeds to step 1903.

  In step 1902, the probe control unit 1607 specifies the ID of the probe 1610 to be controlled, and instructs the management unit 141 to insert or remove the probe 1610. In response to this instruction, the management unit 141 executes insertion or removal of the designated probe 1610. As a result, the monitoring operation of the application program by the probe 1610 is started or stopped. Then, the probe operation control process ends.

  In step 1903, the probe control unit 1607 sets the value of the operation control variable assigned to the probe 1610 to be controlled according to the content of control.

  In step 1904, when the program code of the probe 1610 to be controlled is executed, the operation control code is executed prior to the code for executing the monitoring operation. This operation control code is incorporated in the probe 1610 to be controlled in step 2105 of the flow shown in FIG. The operation control code refers to the operation control variable, determines whether to perform the monitoring operation according to the value, and controls the execution of the code that performs the monitoring operation. Then, the control process of the probe monitoring operation ends.

  In the present embodiment, collection of determination information from the probe 1610 is executed via the collector 1620 of the monitoring event output unit 180a. However, the data acquired by each probe 1610 may be directly output to the condition determination unit 1606 without using the collector 1620. In this case, instead of the setting unit 1609 setting the collector 1620 in step 2006 of the determination information setting process shown in the flow of FIG. 14, the management unit 141 may set the probe 1610. In that case, the management unit 141 uses the probe code generation function to incorporate into the probe a program code that transmits the monitoring item acquired by the probe 1610 directly from the probe to the condition determination unit 1606 through an interface included in the condition determination unit 1606. Execute.

  In the present embodiment, the condition determination unit 1606 and the probe control unit 1607 are both configured as a part of the condition control unit 143. However, all or part of the processing executed by the condition determination unit 1606 or the probe control unit 1607 may be executed by a program code for condition control that the management unit 141 incorporates when generating the program code of the probe 1610.

  In that case, acquisition and transfer of the determination information to the probe 1610 that outputs the determination information and a probe that is a control target of the monitoring operation, condition determination using the determination information and transfer of the determination result, and a probe based on the determination result Program code for executing processing such as control of the monitoring operation is generated in response to an instruction from the setting unit 1609. Then, the generated program code is incorporated into the probe 1610 that outputs the determination information or the probe 1610 to be controlled in the monitoring operation.

  FIG. 17 is an explanatory diagram illustrating an example of the monitoring control setting user interface 110 used to input the probe control definition acquired from the probe control definition storage unit 1602 according to the first embodiment of this invention.

  A monitoring operation control setting window 1850 is displayed on the output device 2711 in order to allow a user (for example, an operation manager or a business manager) to input a probe control definition.

  The monitoring operation control setting window 1850 includes a setting name input field 1852, a control condition input field 1854, a control operation input field 1856, an enter button 1858, and a cancel button 1859.

  The setting name input field 1852 is a field for inputting a name used for management and identification by a user, which is given to the input probe control definition.

  In the control condition input field 1854, a conditional expression for controlling the monitoring operation of the probe 1610 is input. Data used as determination information in the conditional expression is specified by the ID of the probe 1610 and the ID of the data acquisition unit 1630 defined in advance.

  In the control operation input field 1857, an operation for controlling the monitoring operation of the probe 1610, which is executed when the condition input in the control condition input field is satisfied, is input. In the control operation, the ID of the probe 1610 to be controlled in the monitoring operation is designated.

  A button 1855 is a button for displaying an interface for assisting the user in inputting control conditions. A list of operators or functions that can be used to describe the control conditions may be displayed by a user interface (not shown) displayed as a result of operating this button 1855. Furthermore, a list of probes 1610 and data acquisition units 1630 that can be used for determination information may be displayed. The user can select an arbitrary one from the displayed contents. The selected content is reflected in the control condition input field.

  The button 1857 is a button for displaying an interface for assisting the user in inputting a control operation. A list of control operations that can be used to describe the control operation may be displayed by a user interface (not shown) displayed as a result of operating the button 1857. Further, a list of probes 1610 that can be controlled in the monitoring operation may be displayed. The user can select an arbitrary one from the displayed contents. The selected content is reflected in the control operation input field.

  When the setting name, control condition, and control operation are input by the user and the enter button 1858 is operated, the user interface 110 generates a probe control definition according to the content of each input item, and stores it in the probe control definition storage unit 1602. Store.

  Condition control of the monitoring operation of the probe 1610 by the system shown in FIG. 12 can be executed in combination with a system using the user interface shown in FIG. In the system using the user interface shown in FIG. 10, the monitoring setting is executed on the design information, and the probe definition is set based on “monitoring point” and “monitoring data” on the design information designated by the monitoring setting. Generated. Specifically, as already described, the correspondence search unit 115 in FIG. 1 refers to the design correspondence storage unit 130 and executes the processes shown in FIGS. The point where the probe 1610 is inserted in the code and the data to be monitored are specified.

  In this case, in order to specify the probe 1610 as determination information in the user interface shown in FIG. 17 when inputting the probe control definition, the monitoring setting information stored in the monitoring setting table shown in FIG. Can be input by the user. The same applies to the designation of the probe 1610 to be controlled in the monitoring operation. The same ID is assigned to the monitoring setting and the corresponding probe definition. For this reason, the corresponding probe 1610 is specified based on the ID of the designated monitoring setting.

  For example, in order to specify the probe 1610, the probe 1610 to be monitored is set to “a probe that monitors the first argument“ d1 ”of the method“ FuncA.method1 (String d1, String d2) ”” using the probe definition. Even if displayed, a user who does not understand the implementation of the program cannot understand the meaning of the data acquired by the probe 1610. However, if “probe that monitors input data“ a ”in the design information function“ 1 ”” is displayed using the monitoring setting information, the user can easily understand the meaning.

  When setting the control of the monitoring operation of the probe 1610, the user designates the probe 1610 using the monitoring setting information instead of the probe definition information, so that the setting can be made without being aware of the implementation of the application program 170. Can be executed.

  Also, when inputting the control conditions, the monitoring setting is not designated, but the user can input the monitoring point and the monitoring data on the design information using the user interface as shown in FIG. it can. In this case, when the monitoring setting is input on the above design information, the correspondence search unit 115 in FIG. 1 refers to the design correspondence storage unit 130 and determines the probe definition of the probe 1610 corresponding to the input monitoring setting. Create a new one. The generated probe 1610 is used as a probe 1610 for acquiring determination information or a probe 1610 to be controlled in a monitoring operation.

  At this time, it is checked whether or not the same probe definition as the newly generated probe definition already exists in the probe definition storage unit. If the same probe 1610 already exists, that probe 1610 is designated and the existing probe definition is used. On the other hand, if the same probe 1610 does not exist, a probe 1610 for acquiring determination information or a probe 1610 to be controlled for monitoring operation is newly generated according to the newly generated probe definition, and is stored in the application program 170. Inserted.

  Further, in setting the condition control of the probe 1610, when performing the monitoring setting on the design information using the user interface as shown in FIG. 10, the user may simultaneously input the control condition and the control operation. As a result, conditional monitoring settings can be entered. In this case, simultaneously with the generation of the probe definition, a probe control definition is generated and stored in the probe control definition storage unit 1602.

  Next, the configuration and operation of a system that controls the monitoring operation of the probe 1610 according to the probe load will be described. The probe load is a load generated when the probe 1610 is inserted into the application program 170 and the application program 170 is monitored. In order to prevent the probe load from affecting the execution of the application program 170 such as performance degradation, the probe load is measured, and the probe monitoring operation is controlled according to the measured value.

  FIG. 18 is a block diagram illustrating a configuration of a system that controls the monitoring operation of the probe 1610 according to the first embodiment of this invention in accordance with the measured probe load.

  The load control unit 140 includes a measurement unit 2201, a control unit 2203, a setting unit 2205, and a probe log storage unit 2209. The management unit 141 is the same as that shown in FIG.

  The probe log storage unit 2209 is a storage area secured in the main memory 2705 or the external storage device 2707. The measurement unit 2201, the control unit 2203, and the setting unit 2205 are program modules included in the management unit 141. The probe 1610 is a program module inserted into the application program 170 by the management unit 141 shown in FIG.

  First, a process for measuring the probe load will be described.

  When the probe load measurement is instructed, the setting unit 2205 causes the probe 1610 inserted in the application program 170 to send a monitoring operation log of the probe 1610 (hereinafter referred to as a probe log) during the probe load measurement period. Output (S1801).

  The probe log is information related to the monitoring operation executed by the probe 1610, and is output every time the probe 1610 operates and is accumulated in the probe log storage unit 2209. The probe log includes at least information indicating that the probe 1610 has been operated (that is, the probe 1610 has been executed). The probe log may further include various types of information related to processing executed by the probe 1610.

  The process of outputting the probe log may be executed by generating a probe 1610 incorporating a code for causing each probe 1610 to output the probe log, and inserting the generated probe 1610 into the application program 170 again. Alternatively, the probe 1610 may be executed by incorporating a function for outputting a probe log in advance and enabling it.

  Alternatively, when the information necessary for measuring the probe load can be acquired by the monitoring event output from the probe 1610, the monitoring event output unit 180 captures the monitoring event output from each probe 1610 instead of outputting the probe log. Also good. In this case, the monitoring event output unit 180 specifies the operated probe 1610 based on the probe ID added to the monitoring vent, extracts information corresponding to the probe log from the monitoring event, and stores it in the probe log storage unit 2209. Output.

  The measuring unit 2205 calculates the load of each probe 1610 by acquiring the probe log output by each probe 1610 from the probe log storage unit 2209 and counting it. For the calculation of the probe load, the number of operations (that is, the number of executions) of each probe 1610 is used. Hereinafter, a case where the number of probe operations is used will be described as an example. The measuring unit 2205 aggregates the probe logs, calculates the number of operations per unit time of each probe 1610, and uses this value as the probe load.

  When the probe load measurement period ends, the setting unit 2201 causes each probe 1610 to stop outputting the probe log. When the probe log is output by generating the probe 1610 incorporating the code for outputting the probe log, the setting unit 2205 generates the probe 1610 from which the output code of the probe log is removed again. The setting unit 2205 uses the management unit 141 to insert the generated probe 1610 into the application program 170 again.

  With the above processing, the monitoring operation load of each probe 1610 is measured.

  In order to reduce the load of the probe load measurement process itself, the probe log is not temporarily output for each operation of the probe 1610, but each probe 1610 temporarily accumulates the probe log and outputs it at a predetermined interval. Also good. Alternatively, the probe logs temporarily accumulated in the probes 1610 may be aggregated by each probe 1610, and only the statistically processed values may be output at a predetermined interval.

  FIG. 19 is a flowchart of processing in which the load control unit 140 according to the first embodiment of this invention controls the monitoring operation of the probe 1610 according to the measured probe load.

  First, in step 2401, the load control unit 140 starts measuring the probe load. As described above, the probe 1610 incorporating the code for outputting the probe log may be inserted into the application program 170, or the function of outputting the probe log may be validated.

  Next, in step 2402, the load control unit 140 measures the load of the probe 1610.

  Next, in step 2403, the load control unit 140 compares the measured probe load with a preset reference value of the probe load.

  If it is determined in step 2403 that the probe load meets the standard (that is, if the probe load is within the range of the reference value), the process proceeds to step 2407. On the other hand, if it is determined in step 2403 that the probe load does not satisfy the standard, it is necessary to reduce the probe load in order to reduce the influence of the probe load on the execution of the application program 170. Therefore, the process proceeds to step 2405.

  Next, in step 2405, the load control unit 140 controls the probe monitoring operation so as to reduce the probe load.

  Next, in step 2407, the load control unit 140 determines whether or not the set probe load measurement period has ended.

  If it is determined in step 2407 that the measurement period has not expired, the process returns to step 2402 to continue control of probe load measurement and monitoring operations.

  On the other hand, if it is determined in step 2407 that the measurement period has ended, the load control unit 140 ends the probe load measurement in step 2409 and ends the process.

  The reference value of the probe load compared in step 2403 may be set by the absolute value of the load of each probe 1610, or may be set by the relative value with the load of the other probe 1610.

  When using the absolute value of the load, in step 2403, the load control unit 140 compares the load of each probe 1610 with the set reference value, and determines whether or not the load is within the reference range. The reference value of the probe load is stored in advance in the storage area of the load control unit 140.

  On the other hand, when the relative value of the load is used, the load control unit 140 calculates a relative value between the load of the probe 1610 serving as a reference and the load of the probe 1610 to be measured in Step 2403. Then, the load control unit 140 compares the calculated relative value with the set reference value. When the probe load is evaluated using the relative value, a probe 1610 having a particularly high load compared to other probes 1610 can be found regardless of the load of the entire application program 170.

  When there is a probe 1610 whose load exceeds the reference value as a result of comparison with the reference value, the load control unit 140 controls the monitoring operation of the probe 1610. As a method for controlling the monitoring operation, for example, (1) the probe 1610 is removed from the application program 170. (2) The monitoring data output of the probe 1610 is thinned out. Etc.

  According to the method (1), monitoring by the probe 1610 is completely stopped, and the program code of the probe 1610 is removed from the application program 170, so that the influence on the execution of the application program is completely removed.

  On the other hand, according to the method (2), the number of times the monitoring data of the probe 1610 is acquired and output is reduced by, for example, setting the monitoring data of the probe 1610 to be output once every several times. As a result, the load on the probe 1610 is reduced.

  The load control unit 140 measures the probe load again after executing the above control of the monitoring operation.

  When the control (2) is performed during measurement of the monitoring load of the probe 1610, the probe log is always displayed when the probe 1610 is operated regardless of whether the probe 1610 actually outputs monitoring data. By outputting, the probe load can be measured more accurately. The probe log output at this time includes information indicating whether or not the monitoring data is output.

  Alternatively, when the total sum of all probe loads exceeds the reference value, the load control unit 140 adjusts the value of each probe load by the above processing, and controls so that the entire load does not exceed the reference value. it can. In this case, the load control unit 140 calculates the target value of the load of each probe 1610 in consideration of the balance of the load of each probe 1610 and the importance of the monitoring data output from each probe 1610. Control the monitoring operation.

  Measurement and control of the probe load is performed when a probe 1610 is newly added to the application program 170 and when instructed by the user.

  When a new probe 1610 is added to the application 170, the user interface 110 instructs the management unit 141 to insert the new probe 1610 into the application program 170, and at the same time, measures the probe load to the load control unit 140. Instruct.

  The load control unit 140 starts measuring and controlling the load of the probe 1610 including the new probe 1610 in accordance with the insertion of the new probe 1610 executed by the management unit 141 into the application program 170. Then, the load control unit 140 performs measurement and control of the probe load for a preset period, and displays the measured probe load and the control state of the monitoring operation on the user interface 110.

  In addition, the user arbitrarily instructs measurement of the probe load and execution of the control using the user interface 110, and also executes the reference value of the probe load and the control executed when the load exceeds the reference value. The method may be set.

  FIG. 21 is an explanatory diagram illustrating an example of the user interface 110 relating to the measurement of the monitoring operation load according to the first embodiment of this invention.

  Specifically, the user interface 110 in FIG. 21 allows the user to input the probe 1610 to be measured for the monitoring operation load, and further displays the input measurement result of the monitoring operation load of the probe 1610.

  The measurement probe setting window 2600 displays a list of probes 1610 stored in the probe definition storage unit 1601 of the management unit 141. Further, the measurement probe setting window 2600 displays the measured probe load value in the column 2605 when the monitoring operation load of each probe 1610 has already been measured.

  A check box 2601 is provided in each row representing each probe 1610. The user operates the check box 2601 to select the target probe 1610 for measuring the load of the monitoring operation.

  The reference value of the load of each probe is displayed in the reference value input field 2604. The user can input a value in the reference value input field 2604.

  When the user operates the measurement button 2607, the above-described probe load measurement and control are executed, and the measured probe load is displayed in the probe load display field 2605, and the control state of the monitoring operation is displayed in the control state display field 2606. .

  Each probe 1610 or all the probes 1610 is provided with an interface for inputting an allowable reference value of the probe load, and the monitoring operation of the probe with a high load is executed by the flow shown in FIG. 19 together with the measurement of the probe 1610. May be.

  Alternatively, instead of causing each probe 1610 to output a probe load, the probe load may be measured by measuring a performance index of the system or application program 170.

  FIG. 20 is a flowchart of processing in which the load control unit 140 according to the first embodiment of this invention measures the probe load by measuring the performance index of the application program 170.

  In step 2301, the load control unit 140 starts measuring an index indicating the execution performance of the application program 170. The performance index may be measured by a function for measuring the performance index included in the application server 160 or the like. In the present embodiment, the processing throughput of the application program 170 that can be measured by the application server 160 will be described as a performance index used for measuring the probe load.

  Next, in Step 2303, the load control unit 140 measures the processing throughput in a state where the measurement target probe 1610 is not inserted. Specifically, the application program 170 is executed in a state where the measurement target probe 1610 is not inserted into the application program 170, and the average processing throughput for a predetermined period is obtained.

  Next, in step 2304, the probe insertion unit 1608 inserts the measurement target probe 1610 into the application program 170 and executes monitoring by the probe 1610.

  Next, in step 2305, the application program 170 is executed with the measurement target probe 1610 inserted, and the average processing throughput for a predetermined period is measured as in step 2303.

  Next, in step 2306, the load control unit 140 calculates a probe load value using the processing throughput values measured in step 2303 and step 2305.

  Next, in step 2308, the load control unit 140 stops measuring the performance index set in step 2301, and ends the probe load measurement process.

  With the above processing, the probe load of the measurement target probe 1610 is measured.

  The performance index used for measuring the probe load is not limited to the processing throughput. In addition, by measuring a plurality of performance indexes, it is possible to perform more accurate probe load measurement.

  In the flow of FIG. 20, the measurement of the performance index in step 2305 and the calculation of the probe load in step 2306 may be executed simultaneously. Further, the processing after step 2403 in FIG. 19 may be executed using the calculated probe load value. In this case, it is determined whether or not the calculated probe load value is within the range of the reference value, and the monitoring operation of the probe 1610 is controlled as necessary. As a result, in step 2305, when the load of the measurement target probe 1610 is high without waiting for the end of the period for measuring the performance index, the monitoring operation of the measurement target probe 1610 is controlled and the execution of the application program 170 is affected. It can be prevented from occurring.

  In this embodiment, every time the probe load is measured, in step 2303, the performance index is measured in a state where the measurement target probe 1610 is not inserted. Then, in step 2306, the effect of the probe 1610 is calculated using the measured value. However, the performance index measured in the past may be recorded, and the calculation in step 2306 may be executed using the recorded value.

  Further, in the present embodiment, the performance index is measured separately in the state where the measurement target probe 1610 is inserted and in the state where the measurement target probe 1610 is not inserted. However, if possible, both may be measured simultaneously. For example, when the processing time of the method for inserting the measurement target probe 1610 is used as a performance index, the processing time including the processing of the monitoring operation of the measurement target probe 1610 by incorporating the function of the load measurement probe 1610 in the program code of the measurement target probe 1610 Then, the processing time not including the process of the monitoring operation of the measurement target probe 1610 may be measured simultaneously. As a result, both can be measured at the same time, and the load of the probe to be measured can be obtained.

  The first embodiment of the present invention presupposes that one component on the design information is in a design correspondence relationship with only one component of the program code. Hereinafter, as a second embodiment, an example in which a plurality of components of the program code correspond to one component on the design information will be described.

  Hereinafter, only the points of the second embodiment of the present invention different from the first embodiment will be described. The configuration of the second embodiment not described below is the same as that of the first embodiment.

  FIG. 22 is an explanatory diagram illustrating an example of the design correspondence relationship according to the second embodiment of this invention.

  Specifically, FIG. 22 shows an example of design correspondence information in the case where one or a plurality of components of the program code corresponds to one component. When a plurality of constituent elements correspond to one constituent element, the object ID of each corresponding constituent element and the relationship between those constituent elements are stored in the corresponding element 603. The example of FIG. 22 indicates that the process of “function 2” is processed by one of the two methods depending on the situation in the design correspondence 612a.

  Similar to the first embodiment, the correspondence search unit 115 according to the present embodiment executes a search process of the corresponding program code components as shown in FIGS. However, when a design correspondence as shown in the design correspondence 612a exists, the correspondence search unit 115 searches for a plurality of components. The correspondence search unit 115 executes the processing shown in FIGS. 8 and 9 for each searched component, searches for the corresponding component of the program code, and is necessary for realizing the specified monitoring. Generate multiple probe definition sets.

  For example, in the example of FIG. 5, when “A01.Function 2” is specified as a location to be monitored, two methods “FuncB.method2a (Data1 d1)” and method “FuncB.method2b (Data1 d1)” are specified. It is searched as a monitoring location on the program code corresponding to the specified location. As a result, a probe definition corresponding to each location is generated. These probe definitions are given the same probe ID 1001 and a serial number (probe number 1002) for identifying each.

  FIG. 23 is an explanatory diagram illustrating a configuration example of the probe definition storage unit 1601 according to the second embodiment of this invention.

  Specifically, FIG. 23 shows an example of the probe definition when the design correspondence is as shown in FIG. The probes in the second and third lines in FIG. 23 are inserted into the method “FuncB.method2a (Data1 d1)” and the method “FuncB.method2b (Data1 d1)” corresponding to “A01.Function 2” in FIG. 1610 is shown. These probes 1610 have the same probe ID 1001 “P002” and are identified by different probe numbers 1002 “1” and “2”.

  The management unit 141 designates a target probe 1610 using the probe ID 1001, and generates, inserts, and removes the probe 1610 based on all probe definitions having the designated probe ID 1001. In the example shown in FIG. 23, when the management unit 141 instructs insertion of a probe whose probe ID 1001 is “P002”, two probes 1610 are generated and inserted into the application program 170, respectively.

  The following can be cited as representatives of aspects of the present invention other than those described in the claims.

(1) In a computer system including an application server that executes an application program,
The computer system is
Holding an operation executed to control a program module for monitoring the application program, and a condition used to determine whether to execute the operation;
The computer system, wherein the operation is executed when the condition is satisfied.

(2) The computer system
Executing one or more of the program modules that monitor the application program;
The computer system according to (1), wherein whether or not the condition is satisfied is determined based on data acquired by any one of the program modules.

  (3) The operation executed to control the program module that monitors the application program is insertion of the program module into the application program or removal of the program module from the application program. The computer system according to (1).

  (4) The computer system generates a program code for executing all or part of a determination as to whether or not to execute the operation and the operation executed as a result of the determination ( The computer system described in 1).

  (5) The computer system inputs an operation executed for controlling a program module that monitors the application program and a condition used for determining whether to execute the operation. (1) The computer system described in (1) above.

(6) The computer system
Holding information indicating a correspondence relationship between the design information component in one stage of the application program development process and the design information component in a stage advanced from the one stage;
When information indicating a point or data on a business process is input, based on the input information and information indicating the correspondence relationship, a program module that executes the operation, or whether to execute the operation The computer system according to (5), wherein a program module for acquiring data used for determining the data is specified.

(7) The computer system
The computer system according to (6), wherein when the designated program module does not exist, the designated program module is newly generated.

(8) In a computer system including an application server that executes an application program,
The computer system measures a load generated by inserting a program module for monitoring the application program into the application program and executing the program module.

  (9) The computer system according to (8), wherein the computer system measures the load based on a log of a monitoring operation of a program module that monitors the application program.

  (10) The computer system according to (8), wherein the computer system measures the load based on a performance index when the application program is executed.

  (11) The computer system according to (8), wherein the computer system controls an operation of a program module that monitors the application program when the measured load does not satisfy a predetermined condition.

  (12) The computer system controls the operation of the program module by removing the program module that monitors the application program when the measured load does not satisfy a predetermined condition. The computer system according to (11).

(13) The computer system
When inserting a program module for monitoring the application program into the application program, the load by the program module is measured,
The computer system according to (8), wherein the operation of the program module is controlled when the measured load does not satisfy a predetermined condition.

  (14) The computer system controls operation of the program module by removing a program module that monitors the application program when the measured load does not satisfy a predetermined condition. The computer system according to (13).

  (15) The computer system according to (8), wherein the computer system includes a user interface that inputs the program module to be measured for load and the condition and displays the measured load. system.

  By using the present invention, application monitoring can be easily set and executed without being conscious of implementation of the application program.

It is a block diagram which shows the structure of the application program monitoring system of the 1st Embodiment of this invention. It is explanatory drawing of the development process of an application, and the model figure created in the process. It is explanatory drawing which shows the structure of the management part of the 1st Embodiment of this invention. It is explanatory drawing which shows the structural example of the probe definition memory | storage part of the 1st Embodiment of this invention. It is explanatory drawing which shows the code example of the probe produced | generated as advice in the 1st Embodiment of this invention. It is explanatory drawing which shows the example of the program code for adding the probe of the 1st Embodiment of this invention to an application program. It is explanatory drawing which shows the example of the design correspondence of the 1st Embodiment of this invention. It is a flowchart of the process which the correspondence search part of the 1st Embodiment of this invention searches for the point which inserts a probe. It is a flowchart of the process in which the correspondence search part of the 1st Embodiment of this invention analogizes a design correspondence. It is explanatory drawing of the example of the user interface which performs the monitoring setting of an application program on design information in the 1st Embodiment of this invention. It is explanatory drawing which shows the example of the monitoring setting table of the 1st Embodiment of this invention. It is a block diagram which shows the structure of the system which controls the monitoring operation | movement of the probe of the 1st Embodiment of this invention according to conditions. It is explanatory drawing which shows the structural example of the probe control definition memory | storage part of the 1st Embodiment of this invention. It is a flowchart of the process which the setting part of the 1st Embodiment of this invention sets a condition determination part. It is a flowchart of the process which the setting part of the 1st Embodiment of this invention sets a probe control part. It is a flowchart of the process which the probe control part of the 1st Embodiment of this invention controls operation | movement of a probe. It is explanatory drawing which shows the example of the monitoring control setting user interface used in order to input the probe control definition acquired from the probe control definition memory | storage part of the 1st Embodiment of this invention. It is a block diagram which shows the structure of the system which controls the monitoring operation | movement of the probe of the 1st Embodiment of this invention according to the measured probe load. It is a flowchart of the process which the load control part of the 1st Embodiment of this invention controls the monitoring operation | movement of a probe according to the measured probe load. It is a flowchart of the process which the load control part of the 1st Embodiment of this invention measures a probe load by measuring the performance parameter | index of an application program. It is explanatory drawing which shows the example of the user interface regarding the measurement of the monitoring operation | movement load of the 1st Embodiment of this invention. It is explanatory drawing which shows the example of the design correspondence of the 2nd Embodiment of this invention. It is explanatory drawing which shows the structural example of the probe definition memory | storage part of the 2nd Embodiment of this invention. It is a block diagram which shows the physical structure of the application program monitoring system of the 1st Embodiment of this invention.

Explanation of symbols

110 User interface 115 Correspondence search unit 120 Design information search unit 130 Design correspondence storage unit 141 Management unit 143 Condition control unit 145 Load control unit 160 Application server 170 Application program 180, 180a Monitoring event output unit 190 Monitoring event recording unit 550 Design information Display window 551 Design information component 552 Design information 560 Monitoring data selection window 561 Monitoring data selection button 562 Monitoring data list 563 Determination button 901 Monitoring setting ID
903 Monitoring setting name 912 Monitoring point 921 Monitoring data 601 Corresponding element 603 Corresponding element 1001 Probe ID
1002 Probe number 1003 Probe name 1012 Probe insertion point 1021 Monitoring data 1601 Probe definition storage unit 1604 Probe generation unit 1608 Probe insertion unit 1610 Probe 1602 Probe control definition storage unit 1606 Condition determination unit 1607 Probe control unit 1609 Setting unit 1620 Collector 1630 Data acquisition Unit 1801 monitoring control setting ID
1803 Control condition 1805 Control operation 1850 Monitoring operation control setting window 1852 Setting name input field 1854 Control condition input field 1855 Control condition input support interface display button 1856 Control operation input field 1857 Control operation input support interface display button 1858 Decision button 1859 Cancel button 2201 Measurement unit 2203 Control unit 2205 Setting unit 2209 Probe log storage unit 2600 Monitor load measurement setting window 2601 Measurement probe selection check box 2603 Probe name 2604 Load reference value input column 2605 Monitor load display column 2606 Control state display column 2607 Measurement start button 2701 Bus 2703 CPU
2705 Main memory 2707 External storage device 2709 Input device 2711 Output device 2713 Network interface

Claims (18)

In a computer system that executes an application program,
The computer system is
Holding information indicating a correspondence relationship between the design information component in one stage of the application program development process and the design information component in a stage advanced from the one stage;
A computer system characterized in that a component of design information in one stage of the application program development process is associated with a code of the application program based on the information indicating the correspondence relationship. The component of the design information associated with the code of the application program is information indicating a point on the business process,
The computer system according to claim 1, wherein the computer system associates a point on a business process with a point on the code of the application program based on the information indicating the correspondence relationship. When the computer system receives information indicating a point on the business process,
Based on the information indicating the correspondence relationship, a point on the business process identified by the input information is associated with a point on the code of the application program,
3. The computer system according to claim 2, wherein a program module for monitoring the application program is inserted at a point on the code of the associated application program.   4. The computer system according to claim 3, wherein the program module inserted into the application program monitors the application program by acquiring log information of the application program.   5. The computer system according to claim 4, wherein the log information includes data processed by the application program or information indicating a processing time of the application program.   The computer system according to claim 3, further comprising a user interface for inputting information indicating points on the business process. The component of the design information associated with the code of the application program is information indicating data processed in the business process,
The computer system according to claim 1, wherein the computer system associates data processed in a business process with data on a code of the application program based on information indicating the correspondence relationship. When the computer system receives information indicating data to be processed on the business process,
Based on the information indicating the correspondence, the data on the business process identified by the input information is associated with the data on the code of the application program,
The computer system according to claim 7, wherein a program module for acquiring the associated data is generated.   9. The computer system according to claim 8, wherein the data acquired by the program module is log information of the application program.   The computer system according to claim 8, further comprising a user interface for inputting information indicating data processed in the business process.   The components of the design information associated by the computer system include at least activities and connection flows in business process diagrams, messages in UML sequence diagrams, classes, methods and fields in class diagrams, and classes, methods and fields in program code. The computer system according to claim 1, wherein the number is one. The computer system is
A unique identifier in the computer system is assigned to each design information component,
The computer system according to claim 11, wherein information indicating a correspondence relationship between the components of the design information is managed using the identifier. A computer system control method for executing an application program,
The method
Holding information indicating a correspondence relationship between the design information component in one stage of the application program development process and the design information component in a stage advanced from the one stage;
A method of associating a component of design information in one stage of the development process of the application program with a code of the application program based on the information indicating the correspondence relationship. The component of the design information associated with the code of the application program is information indicating a point on the business process,
When the method receives information indicating a point on the business process,
Based on the information indicating the correspondence relationship, a point on the business process identified by the input information is associated with a point on the code of the application program,
14. The method according to claim 13, wherein a program module for monitoring the application program is inserted at a point on the code of the associated application program. The design information associated with the application program code is information indicating data on a business process,
When the method receives information indicating data to be processed on the business process,
Based on the information indicating the correspondence relationship, the data processed on the business process identified by the input information is associated with the data on the code of the application program,
The method according to claim 13, wherein a program module that acquires the associated data is generated. A program for controlling a computer system that executes an application program,
The computer system is
A storage device that stores the program; and a processor that executes the program stored in the storage device.
Holding information indicating a correspondence relationship between the design information component in one stage of the application program development process and the design information component in a stage advanced from the one stage;
The program causes the processor to execute a first procedure for associating a component of design information and a code of the application program in one stage of the application program development process based on the information indicating the correspondence relationship. A program characterized by that. The component of the design information associated with the code of the application program is information indicating a point on the business process,
In the first procedure, when information indicating a point on the business process is input, the point on the business process identified by the input information based on the information indicating the correspondence relationship, and the application Map points on the program code,
The program further causes the processor to execute a second procedure for inserting a program module for monitoring the application program at a point on the code of the associated application program. The listed program. The component of the design information associated with the code of the application program is information indicating data on the business process,
When the information indicating the data to be processed on the business process is input, the first procedure is processed on the business process identified by the input information based on the information indicating the correspondence relationship. And the data on the code of the application program,
The program according to claim 16, further causing the processor to execute a second procedure for generating a program module for acquiring the associated data.

Images